Skip to main content

Long Non-coding RNAs with specific regulation and function in the Drosophila Central Nervous System

Final Report Summary - LNCRNADROCNS (Long Non-coding RNAs with specific regulation and function in the Drosophila Central Nervous System)

The scientific and technical breakthrough of human genome sequencing and the advent of genome wide technologies have dramatically altered our view of genome organization and function. We know now that more than 70% of the human genome is transcribed, although only 2% encodes proteins. Within this huge percentage of supposedly “garbage” transcripts, thousands of non-coding RNAs (ncRNAs) have been described, many of which perform key roles in regulation of developmental processes. The nervous system is one of the most interesting and complex systems. In humans, more than 600 disorders affect central nervous system (CNS) development, causing a wide variety of conditions including motor and psychiatric impairments, and shortening life expectancy. Interestingly, it has been proposed that positive selection of non-coding sequences rather than protein coding genes is primarily responsible for CNS evolution. Interestingly, a novel family of processed ncRNAs (longer than 200 nucleotides, capped and polyadenylated) known as long non-coding RNAs (lncRNAs), has gained the attention of the scientific community because of its involvement in key developmental processes and diseases, including neurodevelopmental, neurodegenerative and neuroimmunological disorders, primary brain tumors, and psychiatric diseases.
Drosophila melanogaster has been used successfully to investigate the molecular and cellular basis of neurodegenerative disorders and complex behaviours such as sleep, learning and memory. With more than a hundred years of genetic research and an exquisite set of tools available, the fruit fly is an ideal model for studying the role of lncRNAs during development of the CNS.
In spite of evidence supporting a role for lncRNAs in CNS development, no systematic analysis has yet been reported. Our primary objective was to perform a detailed survey of lncRNA expression and localization during development in Drosophila CNS and to explore their potential roles using as a model system a small structure essential for memory formation, called the Mushroom Body (MB).
We established a protocol for purifying MB neurons, known as Kenyon Cells (KCs), starting from whole brains. We used this protocol to isolate KCs from L3 larvae and adult flies. In both cases, we used the remaining population of neurons (rest of the brain) as controls. We isolated poly-adenylated RNAs and performed deep sequencing.
This set of samples was complemented by an equivalent set generated by the laboratory of Sean Eddy (Harvard), including RNAseq and ChIPseq data from KCs of adult flies (using a different fly line as well as a different methodology for the sorting) and octopaminergic neurons.
In collaboration with Chris Ponting (Oxford, Edinburgh), one of the leaders in lncRNAs research, we have carried out bioinformatic analyses of the samples and developed a pipeline combining de novo and genome-assisted assemblies, followed by extensive filtering. The overlap between our data and the Kenyon cell data from Sean Eddy’s lab revealed a significant enrichment in both data sets for genes that are indicated as differentially expressed. Most interestingly, we did not observe such an overlap and significance when we compare them to octopaminergic neurons. These results confirmed the robustness of our data.

We identified around 200 new intergenic lncRNAs and almost a thousand that were previously annotated. After applying a series of strict filters, we arrived at a list of 144 lncRNAs.
Finally, we performed a manual curation choosing 44 lncRNAs. Of those, 20 were enriched in MBs of L3 larvae, 16 in MBs from adult brains and 8 were selected because of their high levels of transcription.
To perform a functional characterization of selected lncRNAs we needed to confirm the RNAseq data and, more importantly, to assess the localization and temporal expression of the transcripts. Using fluorescent in situ hybridization (FISH) in Drosophila adult brains has been an extremely challenging task, even with well characterized mRNAs. We tried to solve this problem using different methodologies including the development of new tools of theoretical higher sensitivity. Unfortunately, we were not able to successfully detect RNAs with good reproducibility in our model system.
The results of our project show that during brain development expression of some lncRNAs is regulated in space and time, suggesting that the lncRNAs may have a dedicated role in central nervous system development and/or function. These results are intriguing and open a window of research aimed at understanding their potential function. However, due to tissue complexity and technical limitations, more work is required it to investigate their molecular functions and mechanisms of action. Developing a method that allows the detection of small numbers of RNA molecules in the Drosophila brain is the key next step.